search

machanic / AU_R-CNN

The official implementation code of paper: "AU R-CNN：Encoding Expert Prior Knowledge into R-CNN for Action Unit Detection".
https://arxiv.org/abs/1812.05788
87 stars 24 forks source link

Problem in loading Pretrained Snapshot #8

Open rswethashankar92 opened 4 years ago

rswethashankar92 commented 4 years ago

Hi, I tried to execute demo_AU_RCNN.py, where there is a code to load the pre-trained snapshot using chainers, I end up getting an error,

File "demo_AU_rcnn.py", line 28, in init first_fc_weight = weight_param['head/fc/W'] # shape = (1000, 2048) File "/home/wse1kor/anaconda3/envs/leni/lib/python3.7/site-packages/numpy/lib/npyio.py", line 266, in getitem raise KeyError("%s is not a file in the archive" % key) KeyError: 'head/fc/W is not a file in the archive'

please help me out.

rswethashankar92 commented 4 years ago

Am i supposed to use the vgg16 pretrained model ?

machanic commented 4 years ago

@rswethashankar92 the demo is a visualization demonstration, Can you use python AU_rcnn/train.py --evaluate

swetha-sree commented 4 years ago

Hello @sharpstill using demo_AU_rcnn.py we can visualize the AU unit results?

swetha-sree commented 4 years ago

I am facing the same error, Can you help?

machanic commented 4 years ago

@swetha-sree @rswethashankar92 @dev-srikanth I am hurry for preparing another paper these days, you probably need to provide more detail information? And I will answer you in about 1 month layer. You can fix the bug by yourself.

Alternatively, you can use AU_rcnn/train.py --evaluate to test F1 score, this demo is only to visualize the detection result on the face.

apenzko commented 4 years ago

I am facing the same issue.

@swetha-sree @rswethashankar92 did you find a way to resolve this?

machanic commented 4 years ago

Hi all, Thanks for helping each other, I will convert this code to a clean pytorch version. I hope I can make it better to help you and easy to develop for its extension! Do you have any idea or suggestion of current code? The bug & problem of the demo code will also be fixed then.